Retinol dehydrogenase 10 (Rdh10) catalyzes the first and rate-limiting step in all-trans-retinoic acid (atRA) biosynthesis from retinol by generating retinal, which retinal dehydrogenases convert into atRA. Most Rdh10 research has focused on the visual cycle or embryonic development. Limited work has been done on postnatal Rdh10 function. Yet, atRA controls the fate of mesenchymal stem cell (MSC) differentiation into adipocytes vs osteoblasts in the adult to regulate adiposity and bone health. Mechanisms of atRA action in directing MSC differentiation remain unclear as does regulation of atRA biosynthesis. The literature contains conflicting reports as to the primary target genes regulated by atRA and other differentiation stimuli. Most studies of atRA regulation of the balance between adipocytes vs osteoblasts has been done in cell lines, and has focused on single genes controlling in one stem cell fate, often later during differentiation. The use of cell lines (e.g. 3T3- L1, F442A, C2C12, MC3T3-E1) most likely has contributed to the conflicting conclusions, because cell lines do not faithfully model primary cells, differ from each other, and even cell lines with the same nominal designation often differ as a result of long-term culture. This proposal addresses atRA function in directing MSC fate by proposing study of an in vivo model of atRA insufficiency, which will be accompanied by generating primary mouse embryonic fibroblasts to produce comprehensive mechanistic insight into determining adipocyte vs. osteoblast cell fates. The in vivo model relies on heterozygote ablated Rdh10 (Rdh10+/-) that has a phenotype of increased adiposity and adipocyte proliferation in bone marrow early in life, with decreased calcification of osteoblasts. The Rdh10+/- model (the homozygote is embryonic lethal) provides for consistently attenuated vitamin A function in vivo through reproducibly reduced atRA. This is an innovative model that will provide new insight into atRA function concerning regulating differentiation of adipocytes vs osteoblasts, and new insight into Rdh10 function and atRA biogenesis. Understanding the roles of Rdh10 and atRA in obesity and adipogenesis promises direct human health applications. This expectation is supported by the twenty-six intergenic traits related to body mass index and weight gain that map close to human Rdh10. Aim 1 will determine the phenotype and metabolic consequences of reducing Rdh10 expression and atRA concentrations in vivo (Rdh10+/- mice), focusing on adipocyte and osteoblast differentiation. This aim will test the hypothesis that Rdh10 is a major enzyme that controls atRA homeostasis to regulate retinoid function in adipocyte and osteoblast differentiation. Aim 2 will determine mechanism(s) that underlie the phenotype of Rdh10+/- mice with respect to cell fate determination. This aim will use two sets of MEF: 1) from Rdh10+/- mice; 2) immortalized MEF with a total knockout of Rdh10 to test the hypothesis that atRA generated by Rdh10 directs maturation of MSC by regulating gene expression very early in cell fate determination, and will identify the responsive genes.